Rebuilding Drive Data

1. A system, comprising: an array of disk drives; and a controller communicatively coupled to the array of disk drives, the controller to: identify, within the array of disk drives, a predictive failed disk; rebuild data of the predictive failed disk on a spare drive in a regeneration mode and determine a regeneration time baseline; rebuild data of the predictive failed disk on the spare drive in a copy mode and determine a copy mode execution time; and switch to regeneration mode if the copy mode execution time is greater than the regeneration time baseline.

2. The system of claim 1 , wherein the controller is configured to determine the regeneration time baseline using only samples that have not generated an error.

3. The system of claim 1 , wherein after the switch to regeneration mode, the controller is configured to regenerate a specified number of samples and determine a new regeneration time baseline.

4. The system of claim 1 , wherein upon detecting an unrecoverable error, the controller is configured to switch to regeneration mode and stay in regeneration mode for a predetermined number of samples.

5. The system of claim 1 , wherein the controller is configured to rebuild the data of the predictive failed disk in copy mode for an initial time period after the rebuilding begins.

6. A method, comprising: monitoring a disk array to identify a predictive failed disk; rebuilding data of the predictive failed disk on a spare drive in a regeneration mode and determining a regeneration time baseline; rebuilding data of the predictive failed disk on the spare drive in a copy mode and determining a copy mode execution time; and switching to regeneration mode if the copy mode execution time is greater than the regeneration time baseline.

7. The method of claim 6 , wherein determining the regeneration time baseline comprises determining the regeneration time baseline using only samples that have not generated an error.

8. The method of claim 6 , comprising, after switching to regeneration mode, regenerating a specified number of samples and determining a new regeneration time baseline.

9. The method of claim 6 , comprising, in response to detecting an unrecoverable error, switching to regeneration mode and staying in regeneration mode for a predetermined number of samples.

10. The method of claim 6 , comprising rebuilding the data of the predictive failed disk in copy mode for an initial time period after the rebuilding begins.

11. A tangible, non-transitory, computer-readable medium comprising instructions that direct a processor to: identify a predictive failed disk; rebuild data of the predictive failed disk on a spare drive in a regeneration mode and determine a regeneration time baseline; rebuild data of the predictive failed disk on the spare drive in a copy mode and determine a copy mode execution time; and switch to regeneration mode if the copy mode execution time is greater than the regeneration time baseline.

12. The computer-readable medium of claim 11 , comprising instructions that direct the processor to determine the regeneration time baseline using only samples that have not generated an error.

13. The computer-readable medium of claim 11 , comprising instructions that direct the processor to regenerate a specified number of samples and determine a new regeneration time baseline after switching to regeneration mode.

14. The computer-readable medium of claim 11 , comprising instructions that, in response to detecting an unrecoverable error, direct the processor to switch to regeneration mode and stay in regeneration mode for a predetermined number of samples.

15. The computer-readable medium of claim 11 , comprising instructions that direct the processor to rebuild the data of the predictive failed disk in copy mode for an initial time period after the rebuilding begins.